As Google Glass moves nearer to our eager eyes, we're hearing more about the ways in which we might be able to use the augmented reality specs. The latest? Digitally identifying people by what they're wearing.
The app, named InSight, uses an algorithm that matches people against a pre-taken photo, taking into account the distribution of colours and patterns in their clothing and accessories. A bit like facial-recognition, but on a more complex level.
Don't freak out about privacy just yet, though - InSight can only identify someone if they're wearing the same clothes they were when the reference photo was taken, so there won't be any Minority Report-style human databases getting made.
However, it could be extremely useful in crowded places like airports and festivals. No more "Meet me back by the weird guy in the overalls in ten minutes" when splitting from fellow Glastonbury goers.
Where'd ya get those peepers?
Clothing, hair and jewellery will all be put into the algorithm, which Google Glass will use as a full-body fingerprint to later match against. When identified, the person's name will appear over them, helping you to spot them.
The InSight system has been partly funded by Google, and early tests yielded a 93 per cent success rate, even when identifying people from behind.
Via New Scientist